Skip to content

CON-3166 - Removed rollback for NFS resources when creating NFS volume#437

Open
jlothe wants to merge 1 commit intohpe-storage:v2.5.2-releasefrom
jlothe:CON-3166
Open

CON-3166 - Removed rollback for NFS resources when creating NFS volume#437
jlothe wants to merge 1 commit intohpe-storage:v2.5.2-releasefrom
jlothe:CON-3166

Conversation

@jlothe
Copy link
Copy Markdown
Collaborator

@jlothe jlothe commented Feb 5, 2025

Please review the rollback removal changes when creating NFS volume for 2.5.2 branch

Signed-off-by: jyotsna-lothe <jyotsna.lothe@hpe.com>
Copy link
Copy Markdown
Collaborator

@dileepds dileepds left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

Copy link
Copy Markdown
Contributor

@AnushaY1916 AnushaY1916 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

Copy link
Copy Markdown
Collaborator

@datamattsson datamattsson left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is not working very well.

Assume a customer deploys an NFS server with more memory than is available. The Deployment will be stuck "Pending", the CSI driver is waiting indefinitely for the Deployment status to return "Running" and later create the Service if the Deployment is successful. In this state, if the user decides to delete the frontend PVC realizing the mistake of oversubscribing memory, leaves the backend "Pending" stuck forever and will eventually leave the Deployment in an unmanaged state and manual cleanup is required.

We need to be able to delete "half created" NFS deployments when they're stuck failing.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants